Objectives and Specific Aims: Many neurons throughout the central nervous system receive the majority of their excitatory inputs on protrusions along the dendrites referred to as 'dendritic spines'. In adult animals, synapses occur on an enlarged head of the spine that is connected to the parent dendrite by a constricted neck. Two hypotheses of how spines affect neuronal function will be examined: 1) changes in the dimensions of the spine neck alter its resistance to current flow, thereby providing a simple geometrical mechanism for controlling synaptic efficacy, and 2) the spine neck creates a compartment separating a synapse from the dendrite and from neighboring synapse thereby retaining activation products (e.g., ions or activated proteins) and facilitating local metabolic control of activated synapses. The cytological development of synapses on dendritic spines in area CA1 of the rat hippocampus will compared with reference to maturation of synaptic efficacy, posttetanic potentiation, and longterm potentiation (LTP), a widely studied form of synaptic plasticity thought to be a learning and memory mechanism. The resistance hypothesis for mediating LTP would be supported if constricted spine necks are present at the developmental onset of LTP, and become shorter and wider with the maturation of LTP. The compartmentation hypothesis would be supported by the absence of constricted spine necks at LTP onset and formation of constricted necks with LTP maturation. The progress report with preliminary ultrastructural and modeling results favors the latter hypothesis. A critical physiological test is proposed. Tetanic stimulation that produces LTP will be applied in area CA1 of living hippocampal slices to determine whether spine necks shorten and/or widen (hypothesis 1), or constrict and/or become thinner (hypothesis 2) at two ages when LTP is robust, but spine number and morphology are immature (postnatal days 7 and 15). Health Relatedness: In pathological states that lead to severe mental retardation, alterations in spine morphology can range from complete loss of spines to gross swelling of the spine heads. The proposed study is designed to establish whether spine shape is an important electrotonic influence on synaptic efficacy (hypothesis 1) and/or might act to compartmentize spine synapses (hypothesis 2), in a brain region known to be very sensitive to perinatal disorders, ischemia, and a variety of adult disorders that lead to seizures and/or loss of memory (hippocampal area CA1). Methods: Standard hippocampal slice preparations will be physiologically stimulated to produce and record LTP. Specific areas of the developing neuropil are sampled by electron microscopy of single thin sections combined with identification and measurement through serial sections. Total dimensions of spines, their synapses, and presynaptic varicosities obtained from complete 3-dimensional reconstructions are tested for correlations with each other. Dimensions of spine compartments (head, neck and organelles) are tested with theoretical models to estimate the range in the spine's biophysical properties.